Combined histone deacetylase and NF-κB inhibition sensitizes non-small cell lung cancer to cell death

Combined histone deacetylase and NF-jB inhibition sensitizes non-small cell lung cancer to cell death Brian K. Rundall, DO, Chadrick E. Denlinger, MD, and David R. Jones, MD, Charlottesville, Va

Background. Non-small cell lung cancer (NSCLC) remains resistant to traditional and novel chemotherapeutic agents, relating, in part, to the activation of the antiapoptotic transcription factor NFjB. We hypothesize that inhibition of NF-jB using BAY-11-7085 will sensitize NSCLC cells to death, induced by the histone deacetylase inhibitor suberoylanilide hydroxamic acid (SAHA). Methods. Five tumorigenic NSCLC cell lines (A549, H157, H358, H460, H1299) were treated with nothing, SAHA, BAY-11-7085, or both compounds. Cell death was determined by crystal violet staining. p65 nuclear translocation was determined by Western blot analysis. NF-jB activity was determined by reporter-gene assays and by reverse transcriptase-polymerase chain reaction of the endogenous NFjBedependent gene interleukin 8. Apoptosis was determined by DNA fragmentation. Clonogenic cell survival assays were also performed. Data was analyzed with the Student t test when appropriate. Results. SAHA alone resulted in no meaningful NSCLC cell death. SAHA induced nuclear translocation of p65, which was inhibited by BAY-11-7085. SAHA significantly induced NFjBedependent transcription which was ameliorated after treatment with BAY-11-7085 ( P = .01). Combined SAHA and BAY-11-7085 induced significantly more apoptosis and cell death than either drug alone ( P = .002). Conclusions. Combined HDI and NF-jB inhibition using BAY-11-7085 sensitizes NSCLC cells to cell death and appears promising as a novel treatment strategy for this disease. (Surgery 2004;136:416-25.) From the Department of Surgery, University of Virginia School of Medicine Charlottesville, Va

THIRD-GENERATION COMBINATION chemotherapy regimens used to treat patients with advanced stage non-small cell lung cancer (NSCLC) has recently been shown to have response rates of only 16% to 27%.1 This finding translates into an increase in median survival of only 2 to 3 months.2 These disappointing results, combined with the fact that annual mortality rates for lung cancer are greater than those of breast, colorectal, and prostate carcinomas combined,3 demand that novel treatment strategies for this disease be developed. Increasing evidence indicates that chromatin remodeling through the acetylation and deacetylation of nucleosome core proteins regulates Presented at the 65th Annual Meeting of the Society of University Surgeons, St. Louis, Missouri, February 11-14, 2004. Supported by grants to D.R.J. (NCI CA83920 and the American Association for Cancer Research) and to C.E.D. (NCI F32 CA101497). Reprint requests: David R. Jones, MD, Assistant Professor of Surgery, Department of Surgery, Box 800679, University of Virginia, Charlottesville, VA 22908-0679. 0039-6060/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.surg.2004.05.018

416 SURGERY

transcriptional activation and repression.4,5 Specifically, the acetylation status of nucleosomeassociated histones is regulated by histone acetyltransferases and histone deacetylases. Based on these and other observations there has been an increased interest in the development of histone deacetylase inhibitors (HDIs) as novel agents in the treatment of cancer.4,5 In vitro and in vivo studies have demonstrated a lack of cancer cell proliferation, enhanced cell cycle arrest, and proapoptotic effects after treatment with HDIs.6 While there are several classes of HDIs in various stages of development for the potential treatment of cancer, suberoylanilide hydroxamic acid (SAHA) has shown promise in preclinical studies and is currently in phase I clinical trials.6 We have previously shown that NSCLC cells are resistant to the HDIs sodium butyrate and trichostatin A when used as single agents.7 This lack of significant tumor cell death relates, in part, to HDI-induced activation of the antiapoptotic transcription factor NF-jB. We have subsequently shown that inhibition of NF-jB transcriptional activity with an adenovirally delivered dominant-negative inhibitor of NF-jB dramatically sensitized NSCLC cells to sodium butyrateeinduced apoptosis.7,8

Surgery Volume 136, Number 2

Moreover, we have also shown that combined proteasome and histone deacetylase inhibition with Velcade and either sodium butyrate or SAHA, respectively, enhances NSCLC cell death in vitro.9,10 NF-jB is activated, in part, through the proteasome-mediated degradation of its cytosolic inhibitory protein IjB, which permits translocation of NF-jB to the nucleus where transcription occurs.11 IjB is targeted for ubiquitination and degradation after its phosphorylation by the serine/threonine protein kinase called the IjB kinase or IKK.12 Inhibition of IKK results in sequestration of NF-jB in the cytosol and diminishes or abolishes NF-jBedependent transcription.12 The IKK-b subunit of the IKK complex is approximately 20-fold more active in IjBa phosphorylation than IKK-a as noted in IKK knockout mice studies.13 An increased appreciation of the antiapoptotic role of NF-jB in cancer has generated interest in the design of small molecule inhibitors that block IjB kinase activation. BAY-11-7085, a soluble inhibitor of NF-jB activation, has been shown to inhibit IjB phosphorylation and tumor necrosis factor (TNF)einduced expression of the NF-jBeregulated gene products ICAM1, VCAM, and E-selectin.14 In addition to its inhibition of IjB phosphorylation via IKK and the resultant decrease in DNA-binding function of NFjB,15,16 BAY-11-7085 has also been shown to activate JNK/stresseactivated protein kinase and p38.14 Collectively, these molecular events may have contributed to the recent observations by Scaife et al15 showing that BAY-11-7085 reduced metastatic tumor cell implantation in a colon cancer xenograft model. While we have demonstrated that inhibition of NF-jB sensitizes NSCLC cells to HDI-mediated apoptosis, it is unclear whether inhibition of IKKmediated phosphorylation of IjB will suppress NFjBedependent transcription and sensitize cells to HDI-induced death. Therefore, we hypothesized that inhibition of NF-jB activation using BAY-117085 and inhibition of histone deacetylase activity with the HDI SAHA would result in enhanced cell death in NSCLC cells. MATERIAL AND METHODS Cell culture reagents, and plasmids. Five tumorigenic NSCLC lines (NCI-H157, NCI-H358, NCIH460, NCI-A549 and NCI-H1299) obtained from AmericanTypeCultureCollection(ATCC;Manassas, Va), were cultured in RPMI 1640 (Invitrogen, Carlsbad, Calif) supplemented with 10% fetal bovine

Rundall, Denlinger, and Jones 417

serum (FBS) (HyClone Laboratories, Logan, Utah) and penicillin/streptomycin. The 3x-jB luciferase (3x-jB Luc) reporter construct contains NF-jB DNAebinding consensus sites originally identified in the major histocompatability complex class I promoter, and inserted upstream of Firefly luciferase. The Gal-4 luciferase construct (Gal4-Luc) contains 4 Gal-4 DNA consensus binding sites, derived from the yeast Gal-4 gene promoter, cloned upstream of luciferase complementary DNA (cDNA). The Gal4-p65 fusion protein has the yeast Gal-4 DNAebinding domain fused to full-length p65 (1-551), and was previously described. The dominant-negative IKKb (DN-IKKb) construct has been described previously.17 Antibodies against p65 and RNA pol II were obtained from Upstate Biotechnology (Lake Placid, NY) and Santa Cruz Biotechnology (Santa Cruz, Calif), respectively. Antibodies to p38, phospho-p38, JNK, and phospho-JNK, were obtained from Cell Signaling (Beverly, Mass). Anisomycin was obtained from Sigma-Aldrich (St. Louis, Mo). SAHA was obtained from Biomol (Plymouth Meeting, Pa), and BAY-117085 was obtained from Calbiochem (La Jolla, Calif). Cell viability assays. NSCLC cells at 60% confluency were treated with either nothing or SAHA (5 lmol/L) for 24 hours. Cell viability was then determined by sequentially incubating the cells in 1% glutaraldehyde and 0.5% crystal violet for 15 minutes each. The cells were thoroughly washed and dried, and the crystal violet dye was eluted with Sorenson’s solution (30 mmol/L sodium citrate, 0.02 mol/L HCl, 50% ethanol) at room temperature for 15 minutes. Optical density of the eluent was measured at a wavelength of 570 nm. Luciferase reporter-gene assays. NSCLC cells at 40% to 60% confluency were transfected with either the 3x-jB Luc reporter gene (0.3 lg DNA/ well of 12-well plate) or co-transfected with plasmids coding for the Gal4/p65 fusion protein (0.05 lg/well) and the Gal4/Luc reporter gene (0.25 lg/well) with the use of the Polyfect reagent (Qiagen, Valencia, Calif) according to the manufacturer’s instructions. When appropriate, cells were also transfected with either the DN-IKK plasmid or a vector control (0.7 lg/well). Six hours post-transfection, additional media containing the appropriate pharmacologic agents was added. After treatment for 18 hours, cells were washed with phosphate-buffer saline and then lysed in luciferase reporter buffer (Promega, Madison, Wis). Cells were then sequentially snap frozen at
808C and thawed at 378C. Cell lysates were cleared by

418 Rundall, Denlinger, and Jones

Surgery August 2004

Table. Tumor histologies and molecular characteristics NSCLC line H157 H358 H460 A549 H1299

Tumor histology

p53 status

p16 status

Rb status

Squamous BAC Large cell Unknown Large cell

MUT Null WT WT Null

Null WT Null Null Null

Unknown Null WT WT WT

NSCLC, Non-small cell lung cancer; BAC, bronchioalveolar cell carcinoma; MUT, mutant; WT, wild type.

centrifugation at 13,000g, and protein concentrations were determined with the Pierce BCA protein assay kit (Pierce, Rockford, Ill). Luciferase assays were performed with the substrate DLuciferin; relative light units were measured with an AutoLumat LB953 luminometer (Berthold Analytical Instruments). Luminescence was normalized to protein concentrations. Western blot analysis. NCSLC cells 40% to 50% confluent on 100-mm plates were treated with nothing, SAHA (5 lmol/L), BAY-11-7085 (20 lmol/L), or both compounds for 6 hours. TNF (10 ng/mL) (Sigma-Aldrich) treatment for 15 minutes was used as a positive control. Nuclear extracts were prepared by incubating the pelleted cells in cytoplasmic extract (CE) buffer (0.075% NP-40, 10 mmol/L HEPES, 50 mmol/L KCl, 1 mmol/L EDTA, 1 mmol/L dithiothreitol) containing protease inhibitors at 48C for 6 minutes followed by centrifugation at 1800 rpm. Pelleted nuclei were washed with CE buffer without NP-40 and subsequently lysed in nuclear extract buffer (20 mmol/L Tris, 420 mmol/L NaCl, 1.5 mmol/L MgCl) containing protease inhibitors at 48C for 10 minutes. Nuclear extracts were cleared by centrifugation at 13,000g and protein concentrations determined with the Pierce BCA protein assay kit. Nuclear extracts (30 lg/lane) were resolved by SDS-PAGE and transferred to nitrocellulose membranes. Primary antibodies against p65 and RNA pol II were used for immunoblotting. Wholecell lysates were prepared from similarly treated cells with the use of the RIPA lysis buffer. Proteins (50 lg/lane) were resolved by SDS-PAGE and were immunoblotted with antibodies to p38, phosphop38, JNK, phospho-JNK, and RNA pol II. Anisomycin (25 lg/mL) was used for a positive control in these studies. Reverse transcriptase-polymerase chain reaction. NCSLC cells 40% to 50% confluent on 60-mm plates were treated for 12 hours under previously described experimental conditions. Cells were lysed with Trizol (Invitrogen), and proteins were extracted with chloroform. RNAs

were precipitated with isopropanol and washed with 70% ethanol. cDNAs were created with the use of the Advantage RT for PCR enzyme (Clontech, Palo Alto, Calif) and interleukin 8 (IL-8) cDNA was amplified by PCR with the use of Platinum Taq (Invitrogen) and the primers 59-CTTCCAAGCTGGCCGTGG-39 and 59-TGAATTCTCAGCCTTCTT-39. As a control, GAPDH cDNA was amplified with the use of the primers 59-GTGAGGAGGGGAGATTCAG-39 and 59-GCATCCTGGGCTACACTG-39. PCR products were resolved on a 0.8% agarose gel. Apoptosis assay. NSCLC cells 50% confluent on 12-well plates were treated for 12 hours as described above. Cells were harvested, and apoptosis was quantified by the detection of nucleosomes released into the cytoplasm with the use of the Cell Death Detection ELISA Plus kit (Roche, Indianapolis, Ind) according to the manufacturer’s instructions. Colony formation assay. Thinly plated NSCLC cells were treated as described previously. After a 24hour treatment, cells were washed, and the drugcontaining media was replaced with fresh media. Clonogenic cell survival was determined by allowing cells to grow for an additional week after treatment. Cells were stained with crystal violet as previously described. Surviving colonies were counted in 5, 50-mm2, representative areas. Statistical analysis. Statistical differences between treatment groups were determined by a 2tailed, unpaired Student t test when appropriate. P values < .05 were considered significant. RESULTS Effect of SAHA on NSCLC cell death. Several reports have suggested that SAHA enhances tumor cell death.6 To determine whether NSCLC cells were sensitive to SAHA-induced cell death, we treated several NSCLC cell lines (H157, H358, H460, A549, and H1299) with differing histologies, and the p53, Rb, and p16 expression profiles (Table) with SAHA (5 lmol/L). Importantly, previous

Surgery Volume 136, Number 2

Fig 1. NSCLC cells H157, H358, H460, A549 and H1299 were treated with nothing or SAHA (5 lmol/L) for 24 hours. Cell viability was measured by uptake of the crystal violet dye (*P < .05). The experiment was performed in triplicate and repeated twice.

reports have shown that SAHA effectively inhibits all cellular histone deacetylase activity at a dose of only 0.02 lmol/L.18 As shown in Fig 1, there is only a 20% to 40% reduction in NSCLC cell survival after 24 hours of SAHA treatment. Enhanced cell death did not occur with dose escalation (up to 50 lmol/L, data not shown). Thus, in contrast to other cancers, NSCLC cells are relatively resistant to SAHA-induced cell death. Results from 2 to 4 cell lines are reported in the remainder of the experiments, although similar results were obtained for each cell line. BAY-11-7085 inhibits SAHA-induced nuclear translocation of the NF-jB subunit p65. To determine whether the inability of SAHA to induce cell death in our NSCLC model system was due to activation of the antiapoptotic transcription factor NF-jB, we first performed Western blot analyses on nuclear extracts of NSCLC cells treated with SAHA. As shown in Fig 2, A, the addition of SAHA induced nuclear accumulation of p65, the most transcriptionally active subunit of NF-jB. In addition, pretreatment of these cells with BAY-11-7085 resulted in a significant decrease in SAHA-induced nuclear accumulation of p65 in the H157 and A549 cells, and to a lesser degree in the H358 cells. These studies demonstrate that SAHA induces p65 nuclear translocation and suggest that BAY-117085 is likely functioning as an inhibitor of IKKmediated phosphorylation of IjB. Activation of NF-jB can occur via increased nuclear translocation of p65 as well as by increasing the transactivation potential of nuclear p65.6,17 To determine whether SAHA was also activating the

Rundall, Denlinger, and Jones 419

transactivation potential of p65, we performed luciferase reporter-gene assays using the Gal4/p65 construct. These assays only evaluate the transactivation potential of p65 and are not responsive to events contributing to nuclear translocation of NF-jB. As shown in Fig 2, B, SAHA dramatically induces the transactivation potential of p65, but the addition of BAY-11-7085 to SAHA did not attenuate this effect. This suggests that, in our model system, SAHA activated NF-jB by both inducing its nuclear translocation and by enhancing the transactivation potential of p65. In contrast, the addition of BAY-11-7085 only affects SAHA-induced NF-jB activation by decreasing p65 nuclear translocation, without affecting the transactivation potential of p65. SAHA-induced NF-kBedependent transcription is blocked by BAY-11-7085. While Fig 2, A, demonstrates that SAHA induces nuclear translocation of p65, it does not confirm that this increased nuclear translocation results in enhanced NF-jBedependent transcription. To experimentally address this question, we transiently transfected NSCLC cell lines with the 3x-jB luciferase reporter-gene and luciferase levels assayed. As shown in Fig 3, A, SAHA induced a 4- to 5fold increase in NF-jB transcriptional activity compared to baseline. Importantly, the addition of BAY-11-7085 significantly suppressed by 50% to 70% the SAHA-induced NF-jB activity. Additionally, transient transfections with a dominantnegative inhibitor of IKKb were performed and are shown in Fig 3, B. A similar but more pronounced suppression of SAHA-induced NF-jB activity was observed compared to NF-jB inhibition after treatment with BAY-11-7085. To confirm that SAHA transcriptionally upregulates endogenous NF-jBedependent genes, we performed RT-PCR for the NF-jBeregulated gene IL-8 compared to baseline. This SAHAinduced IL-8 transcriptional upregulation was blocked by the addition of BAY-11-7085 paralleling results seen in the transient transfection assays (Fig 3, A). BAY-11-7085 does not activate phospho-p38 or phospho-JNK. Previous reports have shown that, in addition to inhibiting IKK-mediated phosphorylation of IjB, BAY-11-7085 also activated phosphop38 and JNK. To determine whether this occurred in our system, we treated the H460 and H358 NSCLC cell lines with nothing, SAHA, BAY-11-7085, combined therapy, or anisomycin as a positive control. Western blot analyses for p38, phosphop38, JNK, phospho-JNK, and RNA pol II were performed. As shown in Fig 4, there is no induction

420 Rundall, Denlinger, and Jones

Surgery August 2004

Fig 2. A, Nuclear extracts were prepared from NSCLC cells treated with nothing, SAHA (5 lmol/L), BAY11-7085 (20 lmol/L), or SAHA and BAY-11-7085 for 6 hours. Similar cells were also treated for 15 minutes with TNF (10 ng/mL) as a positive control. Nuclear proteins were analyzed by Western blotting with the use of primary antibodies against p65. Membranes were reprobed with antibodies against RNA pol II as an internal control. The experiment was repeated 3 times. B, NSCLC cells were co-transfected with expression plasmids coding for the Gal4-p65 (1-551) fusion protein and a Gal4-Luc reporter-gene. Six hours posttransfection, cells were treated for 18 hours as described above. Luciferase activity in cell lysates were measured and expressed as the mean + SEM. The experiment was performed in triplicate and repeated twice.

of either kinase after treatment with SAHA or BAY11-7085, or with the combined treatment. Inhibition of NF-jB with BAY-11-7085 sensitizes NSCLC cells to SAHA-induced apoptosis. While the addition of BAY-11-7085 inhibits SAHA-induced NF-jBedependent transcription, the biologic relevance of this, particularly as it relates to cell death, is unknown. To determine whether BAY-11-7085 would sensitize NSCLC cells to undergo apoptosis, we treated the H358 and H460 NSCLC cells with nothing, SAHA (5 lmol/L), BAY-11-7085 (20

lmol/L), or both SAHA and BAY-11-7085. There was significantly more DNA fragmentation in the combined treatment group compared to SAHA or BAY-11-7085 alone (Fig 5). SAHA alone induced significant DNA fragmentation in the H358 cells compared to untreated cells. However, as shown in longer-term cell survival assays (Fig 6), this relative increase in a 12-hour DNA fragmentation assay was not appreciated. To ascertain whether the short-term apoptotic assay described in Fig 5 would be sustained in

Surgery Volume 136, Number 2

Rundall, Denlinger, and Jones 421

Fig 3. A, NSCLC cells were transfected with the 3x-jB reporter gene and were subsequently treated with nothing, SAHA (5 lmol/L), BAY-11-7085 (20 lmol/L), or both compounds at the same doses for 24 hours. B, NSCLC cells were transfected with the 3x-jB reporter gene and either DN-IKK or a vector control. Cells were subsequently treated with nothing or SAHA (5 lmol/L). Luciferase activity was determined and expressed as the mean + SEM. The experiment was performed in triplicate and repeated twice. (*P < .04, yP < .01 vs all other treatment conditions.) C, H358 cells were treated with nothing, SAHA (5 lmol/L), BAY-11-7085 (20 lmol/L), or SAHA and BAY-11-7085 for 12 hours. IL-8 transcription was determined by RT-PCR (performed 3 times).

a longer cell survival assay, the above experimental conditions were repeated in the same cell lines, and colony formation assays were performed. As shown in Fig 6, there are significantly fewer colonies 7 days post-treatment with combined SAHA/BAY-11-7085 than with either drug alone. Interestingly, there was enhanced cell death with BAY-11-7085 alone, which was not appreciated on the shorter time interval apoptotic assays. This may be related to the longer treatment time (24 hours)

with BAY-11-7085 in the clonogenic assays compared to 12 hours in the DNA fragmentation assay. DISCUSSION Molecular-targeted treatment strategies are becoming increasingly attractive to treat lung and other cancers, particularly because of the dismal clinical results of more traditional therapies. Our laboratory and others have focused on molecular

422 Rundall, Denlinger, and Jones

Surgery August 2004

Fig 4. H358 and H460 NSCLC cells were treated as described previously for 1 hour. Whole cell lysates were prepared and analyzed by SDS-PAGE. Immunoblotting was performed with the use of primary antibodies to phospho-p38, p38, phospho-JNK, and JNK. Blots were reprobed for RNA pol II as a loading control. These experimental results were confirmed in 2 separate, identical experiments.

Fig 5. H358 and H460 cells were treated with nothing, SAHA (5 lmol/L), BAY-11-7085 (20 lmol/L), or both SAHA and BAY-11-7085 for 12 hours. Induction of apoptosis was determined by quantification of DNA fragmentation by ELISA. Data represents the mean + SEM of experiments performed in triplicate. (*P < .04 vs all other treatment conditions.)

strategies to alter the tumor cell’s antiapoptotic response to both novel and traditional chemotherapeutic agents.9,19,20 While there remains much enthusiasm for the potential therapeutic role of HDIs in the treatment of cancer,21 this study highlights the important role of HDI-induced activation of the antiapoptotic transcription factor NF-jB as a mechanism of resistance to this therapy. As shown in Fig 1, the ability of SAHA alone to induce cell death is marginal at best. This finding is not unique, as we have observed similar response rates in this NSCLC model system for the structurally dissimilar HDI sodium butyrate as well as for Trichostatin A.7,9 This suggests that, at least in

NSCLC, isolated HDI therapy, even with the more potent HDI SAHA, may have limited clinical utility. Similar to other genotoxic stressors, such as chemotherapy and radiation,19,20 SAHA activates NF-jBedependent transcription. While this is the first paper documenting that SAHA activates NF-jB transcription through enhanced nuclear translocation of p65, other HDIs (including SAHA) have been shown to activate NF-jB by increasing the transactivation potential of p65 (Fig 2, A, B). 7 Of note, BAY-11-7085 inhibited the nuclear translocation of p65 but had no effect on its transactivation potential. We have previously shown that HDIs activate the transactivation potential of p65 through

Surgery Volume 136, Number 2

Rundall, Denlinger, and Jones 423

Fig 6. H358 and H460 NSCLC cells were treated for 24 hours with SAHA (5 lmol/L), BAY-11-7085 (20 lmol/L), or both SAHA and BAY-11-7085 for 24 hours. Cell survival was determined by allowing cells to grow for an additional 7 days after the treatment period. Colonies were counted in 5 representative 50-mm2 areas of the plate. Graphic data represent the mean colony count + SEM. The experiment was repeated twice. (*P < .0002 vs all other treatment conditions.)

an Akt-mediated mechanism involving phosphorylation of the coactivator p300, which allows recruitment of p300 to the p65 promoter region.7 BAY-11-7085 has been shown to enhance phosphorylation of p38 and JNK, as well as regulate the IKK-mediated phosphorylation of IjB.14 As shown in Fig 4, BAY-11-7085 did not induce phosphorylation of p38 or JNK. This observation, together with the data shown in Fig 2, A, suggests that the

mechanism of action of BAY-11-7085 in our model is to inhibit IKK-mediated phosphorylation of IjB. The increased transcriptional activity of NF-jB observed after treatment with SAHA is inhibited by 50% to 75% with the addition of BAY-11-7085 (Fig 3, A, C). Interestingly, BAY-11-7085 does not completely ameliorate NF-jB activity when compared to complete inhibition of IKK with the dominant-negative construct (Fig 3, B). This

424 Rundall, Denlinger, and Jones

suggests that BAY-11-7085 may more selectively block only 1 of the 3 known IKK isoforms or, alternatively, that increasing doses of BAY-11-7085 would have resulted in a more dramatic reduction in NF-jB transcriptional activity. Unfortunately, when the dose of BAY-11-7085 was increased to only 40 lmol/L, it was toxic to the cells (data not shown). BAY-11-7085 was able to significantly inhibit SAHA-induced activation of NF-jB, and, more importantly, it dramatically sensitized all NSCLC cell lines evaluated to SAHA-induced apoptosis and cell death. As shown in Figs 5 and 6, combined SAHA and BAY-11-7085 resulted in more apoptotic cell death than either drug alone in both short- and long-term survival assays. SAHA is known to induce apoptosis through a mitochondrial-mediated process that surprisingly does not require caspase-3 or 8, but instead involves the generation of reactive oxygen species and cleavage of the pro-apoptotic Bcl-2 family member BID.22 Although this study has shown that SAHA-induced activation of the antiapoptotic transcription factor NF-jB can be inhibited by BAY-11-7085 and that combining these 2 agents results in significantly more NSCLC cell death, there are some limitations to the study. First, the dose of SAHA was held constant, and it is possible that increasing the SAHA dose while maintaining the same nontoxic dose of BAY-11-7085 would have resulted in more cell death than we observed. Second, it was noted on the long-term cell survival assay that BAY-11-7085 alone induced death in 50% of the cells. Unfortunately, as described by Hu X et al23 and also noted by our group, higher doses of BAY-11-7085 are toxic and result in significant cellular necrosis. Therefore, treatment with BAY11-7085 as a single therapeutic agent would likely be difficult and less clinically useful. Third, while it is suggested here that this treatment strategy is promising, the in vitro data needs to be confirmed in our NSCLC xenograft model system before more definitive conclusions can be made. Finally, it is possible that the cell lines chosen for these experiments are not representative of NSCLC and that certain combinations of histology plus individual tumor genomic variations would respond differently. CONCLUSION This study has demonstrated that the effectiveness of HDIs, specifically SAHA, in the treatment of NSCLC may be limited secondary to its robust induction of NF-jB transcriptional activation. More-

Surgery August 2004

over, by inhibiting SAHA-induced NF-jB activation with the soluble small molecule inhibitor BAY-117085, NF-jB activity is significantly decreased, and the NSCLC cells are sensitized to SAHAinduced apoptosis and cell death.

REFERENCES 1. Schiller JH, Harrington D, Belani CP, Langer C, Sandler A, Krook J, et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. New Eng J Med 2002;346:92-8. 2. Carney DN. Lung cancer—time to move on from chemotherapy. N Eng J Med 2002;346:126-8. 3. Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Cancer Statistics 2003. CA Cancer J Clin 2003;53:5-26. 4. Marks PA, Richon VM, Breslow R, Rifkind RA. Histone deacetylase inhibitors as new cancer drugs. Curr Opin Oncol 2001;12:477-83. 5. Johnstone RW. Histone-deacetylase inhibitors: Novel drugs for the treatment of cancer. Nat Rev Drug Discov 2002;1: 287-99. 6. Marks PA, Richon VM, Rifkind RA. Histone deacetylase inhibitors: Inducers of differentiation or apoptosis of transformed cells. J Natl Cancer Inst 2000;92:1210-6. 7. Mayo MW, Denlinger CE, Broad RM, Yeung F, Reily ET, Shi Y, et al. Ineffectiveness of histone deacetylase inhibitors to induce apoptosis involves the transcriptional activation of NF-kappa B through the Akt pathway. J Biol Chem 2003; 278:18980-9. 8. Denlinger CE, Broad RM, Reilly E, Mayo MW, Jones DR. Inhibition of histone deacetylation and NF-jB activation sensitizes NSCLC to undergo apoptosis. J Am Col Surg 2002;195:S27. 9. Denlinger CE, Keller MD, Mayo MW, Broad RM, Jones DR. Combined proteasome and histone deacetylase inhibition in non-small cell lung cancer. J Thorac Cardiovasc Surg 2004;127:1078-86. 10. Denlinger CE, Mayo MW, Jones DR. Proteasome inhibition sensitizes NSCLC to generation of reactive oxygen species and cell death. Lung Cancer 2003;41:S177. 11. Baldwin AS. The NF-kappa B and I kappa B proteins: new discoveries and insights. Annu Rev Immunol 1996;14: 649-83. 12. Karin M, Delhase M. The IjB kinase (IKK) and NF-jB: key elements of proinflammatory signaling. Semin Immunol 2000;12:85-98. 13. Prajapati S, Verma U, Yamamoto Y, Kwak Y, Gaynor R. Protein Phosphatase 2Cb association with the IjB kinase complex is involved in regulating NF-jB activity. J Biol Chem 2004;279:1739-46. 14. Pierce JW, Schoenleber R, Jesmok G, Best J, Moore SA, Collins T. Novel inhibitors of cytokine-induced IjBa phosphorylation and endothelial cell adhesion molecule expression show anti-inflammatory effects in vivo. J Biol Chem 1997;272:21096-103. 15. Scaife CL, Kuang J, Wills JC, Trowbridge B, Gray P, Manning BM, et al. Nuclear factor jB inhibitors induce adhesion-dependent colon cancer apoptosis: implications for metastasis. Cancer Res 2002;62:6870-8. 16. Pham PV, Tamayo AT, Yoshimura LC, Lo P, Ford RJ. Inhibition of constitutive NF-jB activation in mantle cell lymphoma B cells leads to induction of cell cycle arrest and apoptosis. J Immunol 2003;171:88-95.

Surgery Volume 136, Number 2

17. Madrid LV, Mayo MW, Reuther JY, Baldwin AS. Akt stimulates the transactivation potential of the Rel A/p65 subunit of NF-jB through utilization of the IjB kinase and activation of the mitogen-activated protein kinase p38. J Biol Chem 2001;276:18934-40. 18. Richon VM, Emiliant S, Verdin E, Webb Y, Breslow R, Rifkind RA, et al. A class of hybrid polar inducers of transformed cell differentiation inhibits histone deacetylases. Proc Natl Acad Sci 1998;95:3003-7. 19. Jones DR, Broad RM, Madrid LW, Baldwin AS, Mayo MW. Inhibition of NF-jB sensitizes non-small cell lung cancer cells to chemotherapy induced apoptosis. Ann Thorac Surg 2000;70:930-7. 20. Jones DR, Broad RM, Comeau LD, Parsons SJ, Mayo MW. Inhibition of nuclear factor jB chemosensitizes non-small

Rundall, Denlinger, and Jones 425

cell lung cancer through cytochrome c release and caspase activation. J Thorac Cardiovasc Surg 2002;123:310-7. 21. Vigushin DM, Coombes RC. Histone deacetylase inhibitors in cancer treatment. Anti Cancer Drugs 2002;13:1-13. 22. Ruefli AA, Ausserlechner MJ, Bernard D, Sutton VR, Tainton KM, Kohler R, et al. The histone deacetylase inhibitor and chemotherapeutic agent suberoylanilide hydroxamic acid (SAHA) induces a cell death pathway characterized by cleavage of BID and production of reactive oxygen species. Proc Natl Acad Sci U S A 2001;98: 10833-8. 23. Hu X, Janssen W, Moscinski L, Bryington M, Dangsupa A, Rezai-Zadeh N, et al. An IjB inhibitor causes leukemia cell death through a p38 MAP kinase-dependent, NF-jB-independent mechanism. Cancer Res 2001;61: 6290-6.